0000312: n's displayed as y's in Adobe Reader

(0000786)
Alister
04-03-09 05:36

Wait... I can always reproduce it. I've uploaded the source files now.
If I start with 2.pdf, add 1.pdf, then 3.pdf, the first page is displayed wrong.
If I add 3.pdf then 1.pdf it is displayed OK.
I think there is something about 3.pdf that pdfedit doesn't handle quite right.

BTW I've only tried the page order 1,2,3

(0000789)
hockm0bm
04-03-09 09:27

Thanks for report, I will try to look at it as soon as I am done with other issues.

(0000799)
hockm0bm
05-14-09 11:34

Hi, sorry for such a long time without any response, but I was busy with another stuff.

I am not sure I have understood your reproduction steps. What I have tried is the following:
* open 2.pdf
* insert all pages (3) from 1.pdf (before the first page in the original document)
* insert one page from 3.pdf (also at the beginning)
* save
* open in acrobat reader and everything seems to be OK. (first page is a map, then 3 pages of Apendix - B, C, D and tables on the last 2 pages).

I haven't observed neither any n->y nor "Re port".

Are my reproduction steps correct?

(0000809)
Alister
05-15-09 05:46

> * insert one page from 3.pdf (also at the beginning)

No; it only happened if I inserted 3.pdf at the end.
But I can't reproduce it now anyway. The only possible reasons I can think of are:
- a change in my Windows or my Cygwin environment or the pdfedit setup (I have definitely changed pdfedit settings).
- a broken file got fixed by uploading it here (that seems pretty unlikely).

(0000810)
Alister
05-15-09 05:58

Have you looked at "Appendix B.pdf"?
It still displays incorrectly for me in Adobe Reader, but correctly in pdfedit, so maybe we can figure out why...

(0000811)
Alister
05-15-09 06:15

FWIW it also displays incorrectly in GSview, AutoCAD and PDFfill, but correctly in Okular and foxit reader :)

(0000825)
hockm0bm
05-18-09 14:45

I have looked closer to the document and it seems that the problem is caused by any change done to the page inserted from document 1.pdf (this one is inserted to the Appendix B.pdf).

If you save revision 1 from Appendix B.pdf and display it in acrobat reader then you get proper result. However, when you try to do any change to the 1st page then you get n->y problem.

I have tried to modify the original 1.pdf document and there is the very same result. Each modified page (I have drawn a line on the 1st and 3rd page) is affected while the untouched one is correct.

One thing which is not clear to me is that whenever we do some modification to the page the original content stream is not touched and we simply add another content stream (Is this right, Jozo?). But if this is true then how could we result in different rendering?

(0000826)
hockm0bm
05-18-09 14:55

I have just noticed that [7 0] (original content stream of the page 1 - 1.pdf) is changed after I have added an arrow to the page 1 [*].
The arrow itself is represented as a separate content stream though. The page refers both content streams via /Contents.
Why is the original stream changed?

Original document (page 1):
[4 0]:
<<
/Type /Page
/Contents 7 0 R
/Parent 2 0 R
/Resources <<
/ExtGState 8 0 R
/XObject 9 0 R
/ProcSet [ /PDF /ImageC /Text ]
/Font 10 0 R
>>
/Rotate 0
/MediaBox [ 0 0 595 842 ]
>>

Changed document (page 1):
[4 0]:
<<
/Type /Page
/Contents [ 7 0 R 26 0 R ]
/Parent 2 0 R
/Resources <<
/ExtGState 8 0 R
/XObject 9 0 R
/ProcSet [ /PDF /ImageC /Text ]
/Font 10 0 R
>>
/Rotate 0
/MediaBox [ 0 0 595 842 ]
/CropBox [ 0 0 595 842 ]
>>

[*] Here is the xref table for the changed document ()
xref
4 1
0000027988 00000 n
7 1
0000028218 00000 n
26 1
0000030055 00000 n
trailer

(0000827)
hockm0bm
05-18-09 14:58

7_0.orig original content stream
7_0.changed content stream from the changed page

I have tried to compare them, but formating seems to be different, so I couldn't see much from it.

(0000835)
Alister
05-20-09 07:09
edited on: 05-20-09 07:17

I've got another pdf with a similar problem in an inserted page (now running cvs from a few days ago). Every o on the page has been changed to r, and the one h in the title has been changed to d.
Unfortunately I'm getting the same error (freezing the browser) trying to upload it, even though it is well under the 8MB limit and I haven't zipped it or anything.
Do you want me to send you it as well (and the original files)?

Application error 401
Database query failed. Error received from database was 1153: Got a packet bigger than 'max_allowed_packet' for the query: INSERT INTO mantis_bug_file_table
(bug_id, title, description, diskfile, filename, folder, filesize, file_type, date_added, content)
VALUES
(312, ", ", 'db74fca05d17db08f7bc...

(0000836)
Alister
05-20-09 07:14

FWIW this latest pdf came from MS Word 2003, via pdf995.
The affected page in the last example also came from Word, I think via PDFill's pdf printer (but I'm not sure - it could have also been pdf995). I think both these pdf printers use Ghostscript.

(0000838)
Alister
05-20-09 08:15

Also, I tried flattening it, just to see what would happen, and it still displays correctly in pdfedit and incorrectly in Adobe Reader.

(0000839)
hockm0bm
05-20-09 09:13

> Do you want me to send you it as well (and the original files)?

Yes, please.

(0000840)
Alister
05-20-09 09:28

OK, you should receive them soon.

> Unfortunately I'm getting the same error
Sorry if I confused any one... the same error as I mentioned in http://pdfedit.petricek.net/bt/view.php?id=313 [^]

(0000891)
misuj1am
07-13-09 00:44

as this needs GUI, waiting for q4 again

(0000936)
misuj1am
08-13-09 02:02
edited on: 08-13-09 02:02

i created content stream text files (1,2) on win32 and there is a clear difference.

michal, if you have time, can you look why manipulating pages changes also content stream?

(0000955)
hockm0bm
08-20-09 13:02

> michal, if you have time, can you look why manipulating pages changes also content stream?

I am not sure what you mean... Something like http://pdfedit.petricek.net/bt/view.php?id=328? [^]

(0001062)
hockm0bm
05-03-10 21:23

We believe that this is the same problem as bt#348 which has a fix in the CVS.

Issue History
Date Modified	Username	Field	Change
04-03-09 05:07	Alister	New Issue
04-03-09 05:07	Alister	File Added: Appendix B.pdf
04-03-09 05:33	Alister	File Added: 1.pdf
04-03-09 05:33	Alister	File Added: 2.pdf
04-03-09 05:34	Alister	File Added: 3.pdf
04-03-09 05:36	Alister	Note Added: 0000786
04-03-09 09:26	hockm0bm	Status	new => assigned
04-03-09 09:26	hockm0bm	Assigned To	=> hockm0bm
04-03-09 09:27	hockm0bm	Note Added: 0000789
05-14-09 11:34	hockm0bm	Note Added: 0000799
05-15-09 05:46	Alister	Note Added: 0000809
05-15-09 05:58	Alister	Note Added: 0000810
05-15-09 06:15	Alister	Note Added: 0000811
05-18-09 14:44	hockm0bm	Assigned To	hockm0bm => misuj1am
05-18-09 14:45	hockm0bm	Note Added: 0000825
05-18-09 14:55	hockm0bm	Note Added: 0000826
05-18-09 14:57	hockm0bm	File Added: 7_0.orig
05-18-09 14:57	hockm0bm	File Added: 7_0.changed
05-18-09 14:58	hockm0bm	Note Added: 0000827
05-20-09 07:09	Alister	Note Added: 0000835
05-20-09 07:14	Alister	Note Added: 0000836
05-20-09 07:17	Alister	Note Edited: 0000835
05-20-09 08:15	Alister	Note Added: 0000838
05-20-09 09:13	hockm0bm	Note Added: 0000839
05-20-09 09:28	Alister	Note Added: 0000840
07-13-09 00:44	misuj1am	Note Added: 0000891
08-13-09 02:00	misuj1am	File Added: 1
08-13-09 02:01	misuj1am	File Added: 2
08-13-09 02:02	misuj1am	Note Added: 0000936
08-13-09 02:02	misuj1am	Note Edited: 0000936
08-13-09 02:02	misuj1am	Note Edited: 0000936
08-20-09 13:02	hockm0bm	Note Added: 0000955
08-21-09 17:48	hockm0bm	Issue Monitored: hockm0bm
04-30-10 11:22	hockm0bm	Relationship added	duplicate of 0000348
05-03-10 21:23	hockm0bm	Duplicate ID	0 => 348
05-03-10 21:23	hockm0bm	Status	assigned => resolved
05-03-10 21:23	hockm0bm	Resolution	open => duplicate
05-03-10 21:23	hockm0bm	Note Added: 0001062
09-29-10 08:19	misuj1am	File Deleted: Appendix B.pdf
09-29-10 08:19	misuj1am	File Deleted: 1.pdf
09-29-10 08:19	misuj1am	File Deleted: 3.pdf
09-29-10 08:19	misuj1am	File Deleted: 2.pdf
09-29-10 08:19	misuj1am	File Deleted: 7_0.orig
09-29-10 08:19	misuj1am	File Deleted: 7_0.changed
09-29-10 08:19	misuj1am	File Deleted: 1
09-29-10 08:19	misuj1am	File Deleted: 2

Notes
(0000786) Alister 04-03-09 05:36	Wait... I can always reproduce it. I've uploaded the source files now. If I start with 2.pdf, add 1.pdf, then 3.pdf, the first page is displayed wrong. If I add 3.pdf then 1.pdf it is displayed OK. I think there is something about 3.pdf that pdfedit doesn't handle quite right. BTW I've only tried the page order 1,2,3

(0000789) hockm0bm 04-03-09 09:27	Thanks for report, I will try to look at it as soon as I am done with other issues.

(0000799) hockm0bm 05-14-09 11:34	Hi, sorry for such a long time without any response, but I was busy with another stuff. I am not sure I have understood your reproduction steps. What I have tried is the following: * open 2.pdf * insert all pages (3) from 1.pdf (before the first page in the original document) * insert one page from 3.pdf (also at the beginning) * save * open in acrobat reader and everything seems to be OK. (first page is a map, then 3 pages of Apendix - B, C, D and tables on the last 2 pages). I haven't observed neither any n->y nor "Re port". Are my reproduction steps correct?

(0000809) Alister 05-15-09 05:46	> * insert one page from 3.pdf (also at the beginning) No; it only happened if I inserted 3.pdf at the end. But I can't reproduce it now anyway. The only possible reasons I can think of are: - a change in my Windows or my Cygwin environment or the pdfedit setup (I have definitely changed pdfedit settings). - a broken file got fixed by uploading it here (that seems pretty unlikely).

(0000810) Alister 05-15-09 05:58	Have you looked at "Appendix B.pdf"? It still displays incorrectly for me in Adobe Reader, but correctly in pdfedit, so maybe we can figure out why...

(0000811) Alister 05-15-09 06:15	FWIW it also displays incorrectly in GSview, AutoCAD and PDFfill, but correctly in Okular and foxit reader :)

(0000825) hockm0bm 05-18-09 14:45	I have looked closer to the document and it seems that the problem is caused by any change done to the page inserted from document 1.pdf (this one is inserted to the Appendix B.pdf). If you save revision 1 from Appendix B.pdf and display it in acrobat reader then you get proper result. However, when you try to do any change to the 1st page then you get n->y problem. I have tried to modify the original 1.pdf document and there is the very same result. Each modified page (I have drawn a line on the 1st and 3rd page) is affected while the untouched one is correct. One thing which is not clear to me is that whenever we do some modification to the page the original content stream is not touched and we simply add another content stream (Is this right, Jozo?). But if this is true then how could we result in different rendering?

(0000826) hockm0bm 05-18-09 14:55	I have just noticed that [7 0] (original content stream of the page 1 - 1.pdf) is changed after I have added an arrow to the page 1 []. The arrow itself is represented as a separate content stream though. The page refers both content streams via /Contents. Why is the original stream changed? Original document (page 1): [4 0]: << /Type /Page /Contents 7 0 R /Parent 2 0 R /Resources << /ExtGState 8 0 R /XObject 9 0 R /ProcSet [ /PDF /ImageC /Text ] /Font 10 0 R >> /Rotate 0 /MediaBox [ 0 0 595 842 ] >> Changed document (page 1): [4 0]: << /Type /Page /Contents [ 7 0 R 26 0 R ] /Parent 2 0 R /Resources << /ExtGState 8 0 R /XObject 9 0 R /ProcSet [ /PDF /ImageC /Text ] /Font 10 0 R >> /Rotate 0 /MediaBox [ 0 0 595 842 ] /CropBox [ 0 0 595 842 ] >> [] Here is the xref table for the changed document () xref 4 1 0000027988 00000 n 7 1 0000028218 00000 n 26 1 0000030055 00000 n trailer

(0000827) hockm0bm 05-18-09 14:58	7_0.orig original content stream 7_0.changed content stream from the changed page I have tried to compare them, but formating seems to be different, so I couldn't see much from it.

(0000835) Alister 05-20-09 07:09 edited on: 05-20-09 07:17	I've got another pdf with a similar problem in an inserted page (now running cvs from a few days ago). Every o on the page has been changed to r, and the one h in the title has been changed to d. Unfortunately I'm getting the same error (freezing the browser) trying to upload it, even though it is well under the 8MB limit and I haven't zipped it or anything. Do you want me to send you it as well (and the original files)? Application error 401 Database query failed. Error received from database was 1153: Got a packet bigger than 'max_allowed_packet' for the query: INSERT INTO mantis_bug_file_table (bug_id, title, description, diskfile, filename, folder, filesize, file_type, date_added, content) VALUES (312, ", ", 'db74fca05d17db08f7bc...

(0000836) Alister 05-20-09 07:14	FWIW this latest pdf came from MS Word 2003, via pdf995. The affected page in the last example also came from Word, I think via PDFill's pdf printer (but I'm not sure - it could have also been pdf995). I think both these pdf printers use Ghostscript.

(0000838) Alister 05-20-09 08:15	Also, I tried flattening it, just to see what would happen, and it still displays correctly in pdfedit and incorrectly in Adobe Reader.

(0000839) hockm0bm 05-20-09 09:13	> Do you want me to send you it as well (and the original files)? Yes, please.

(0000840) Alister 05-20-09 09:28	OK, you should receive them soon. > Unfortunately I'm getting the same error Sorry if I confused any one... the same error as I mentioned in http://pdfedit.petricek.net/bt/view.php?id=313 [^]

(0000891) misuj1am 07-13-09 00:44	as this needs GUI, waiting for q4 again

(0000936) misuj1am 08-13-09 02:02 edited on: 08-13-09 02:02	i created content stream text files (1,2) on win32 and there is a clear difference. michal, if you have time, can you look why manipulating pages changes also content stream?

(0000955) hockm0bm 08-20-09 13:02	> michal, if you have time, can you look why manipulating pages changes also content stream? I am not sure what you mean... Something like http://pdfedit.petricek.net/bt/view.php?id=328? [^]

(0001062) hockm0bm 05-03-10 21:23	We believe that this is the same problem as bt#348 which has a fix in the CVS.

PDFedit	Bugtracker